Advertisement

Research on the Application of Instance Segmentation Algorithm in the Counting of Metro Waiting Population

  • Yan Cang
  • Chan ChenEmail author
  • Yulong Qiao
Conference paper
  • 24 Downloads
Part of the Advances in Intelligent Systems and Computing book series (AISC, volume 1107)

Abstract

With the development of deep learning, intelligent security and measurement and control products emerge in endlessly. In order to maintain the operation order of subway station, it is one of the important problems in the field of deep learning to carry out real-time and accurate counting of the number of people queued in front of the subway. In order to solve this problem, we propose a simple and flexible real-time counting method for the number of people queued in front of the subway. The image is collected by the camera in front of the subway door, instance segmentation algorithm accurately divides the target in the image, and completes the counting of the number of people queued in front of the subway by calculating the number of targets segmented. Selected the mainstream Mask R-CNN as the basic algorithm, combine the characteristics of the live picture, feature pyramid network and non-maximum suppression process of Mask R-CNN are improved. The experimental results show that the algorithm can realize accurate and real-time counting of the number of people queued in front of the subway, and the accuracy of counting can reach 96%. Compared with the traditional target detection algorithm, it has stronger adaptability to occlusion problem, and can accurately segment the intersection of targets from less than 30% to less than 60%. And the real-time performance has been greatly improved, the target inventory in a single picture only needs about 0.2 s.

Keywords

Instance segmentation Target counting Deep learning Feature pyramid network 

References

  1. 1.
    Hariharan, B., Arbeláez, P., Girshick, R., et al.: Simultaneous Detection and Segmentation (2014)Google Scholar
  2. 2.
    Hariharan, B., Arbeláez, P., Girshick, R., et al.: Hypercolumns for object segmentation and fine-grained localization. In: 2015 IEEE Conference on Computer Vision and Pattern Recognition (CVPR). IEEE (2015)Google Scholar
  3. 3.
    Dai, J., He, K., Sun, J.: Convolutional Feature Masking for Joint Object and Stuff Segmentation (2014)Google Scholar
  4. 4.
    Pinheiro, P.O., Collobert, R., Dollar, P.: Learning to Segment Object Candidates (2015)Google Scholar
  5. 5.
    Pinheiro, P.O., Lin, T.Y., Collobert, R., et al.: Learning to Refine Object Segments (2016)Google Scholar
  6. 6.
    Zagoruyko, S., Lerer, A., Lin, T.Y., et al.: A MultiPath Network for Object Detection (2016)Google Scholar
  7. 7.
    Dai, J., He, K., Sun, J.: Instance-aware Semantic Segmentation via Multi-task Network Cascades (2015)Google Scholar
  8. 8.
    Li, Y., Qi, H., Dai, J., et al.: Fully Convolutional Instance-aware Semantic Segmentation (2016)Google Scholar
  9. 9.
    Ren, S., He, K., Girshick, R., et al.: Faster R-CNN: towards real-time object detection with region proposal networks. IEEE Trans. Pattern Anal. Mach. Intell. 39(6), 1137–1149 (2015)CrossRefGoogle Scholar
  10. 10.
    He, K., Gkioxari, G., Dollar, P., et al.: Mask R-CNN. IEEE Trans. Pattern Anal. Mach. Intell. (2017)Google Scholar
  11. 11.
    Gidaris, S., Komodakis, N.: Object detection via a multi-region and semantic segmentation-aware CNN model. In: IEEE International Conference on Computer Vision, pp. 1134–1142. IEEE (2016)Google Scholar
  12. 12.
    Montserrat, D.M., Lin, Q., Allebach, J., et al.: Training object detection and recognition CNN models using data augmentation. Electron. Imaging 2017(10), 27–36 (2017)CrossRefGoogle Scholar
  13. 13.
    Bodla, N., Singh, B., Chellappa, R., et al.: Soft-NMS—Improving Object Detection with One Line of Code (2017)Google Scholar
  14. 14.
    He, Y., Zhu, C., Wang, J., et al.: Bounding Box Regression with Uncertainty for Accurate Object Detection (2018)Google Scholar
  15. 15.
    Singh, B., Davis, L.S.: An Analysis of Scale Invariance in Object Detection - SNIP (2017)Google Scholar
  16. 16.
    Cai, Z., Vasconcelos, N.: Cascade R-CNN: Delving into High Quality Object Detection (2017)Google Scholar
  17. 17.
    Cholakkal, H., Sun, G., Khan, F.S., et al.: Object Counting and Instance Segmentation with Image-level Supervision (2019)Google Scholar
  18. 18.
    Han, C., Murao, K., Satoh, S., et al.: Learning More with Less: GAN-based Medical Image Augmentation (2019)Google Scholar
  19. 19.
    Perez, L., Wang, J.: The Effectiveness of Data Augmentation in Image Classification using Deep Learning (2017)Google Scholar
  20. 20.
    Chen, X., Girshick, R., He, K., et al.: TensorMask: A Foundation for Dense Object Segmentation (2019)Google Scholar
  21. 21.
    Liu, S., Qi, L., Qin, H., et al.: Path Aggregation Network for Instance Segmentation (2018)Google Scholar

Copyright information

© Springer Nature Singapore Pte Ltd. 2020

Authors and Affiliations

  1. 1.Harbin Engineering UniversityHarbinChina

Personalised recommendations